Components Of A Simple Biometric System

Published Date: 02 Nov 2017

ABSTRACT

Biometrics is the study of automatically recognition by physical or behavioral characteristics of human, such as fingerprints, face, hand geometry, Palm, iris recognition, DNA, typing Rhythm, voice, and iris Gait, In this paper, the main focus is on only five of the different biometrics techniques like fingerprints, hand geometry, iris recognition face and voice recognition then the focus will be on Finger Prints which will include the design of the database and interfaces.

1. Introduction:

Biometrics �refers to the automatic recognition of individuals based on their [unique] physiological and/or behavioral characteristics�[1] and in computing is usually used for security applications that focus on either verification (Is this person who they claim to be?) or identification (Who is this person?). Whilst in the medical profession biometrics refers to collecting the �measurements of biological and/or medical phenomena�[2], for example, measuring a patient�s blood pressure.

Authentication is the process designed to verify a user's identity. The goal of authentication is to protect a system against unauthorized use. This feature enables also the protection of subscribers by denying the possibility for intruders to impersonate authorized users. Authentication procedures are based on the following premises:

� Proof by Knowledge: what the person knows.

� Proof by Possession: what the person owns.

� Proof by Property: what the person is.

Traditional technologies (based on the first two premises) are not sufficient to reduce the impact of counterfeiting. Divisional convenient security barriers are needed as our society gets more and more computer Dependent.

Automatic human identification has become an important issue in today's information and network-based society. The techniques for automatically Profile Biometry 2006 Chapter 1 Introduction Identifying an individual based on his physiological characteristics is called biometrics, which provides an answer to this need.

Biometric identification is, simply, the technique of verifying a person by a physical characteristic or personal trait. Our brains perform biometrics in

distinguishing our associates from our family although, occasionally, there may be similarities.

The traditional and nonsubject specific definition will be used which is �composed of the two Greek terms �bios� for life and �metros� for metric.� [2]. Put simply, the measurement of life, or more specifically, the measurement of human life.

?

2. History of Biometrics:

The first recorded evidence of using biometric authentication was in ancient Egypt. One of the administrators , during the construction of great pyramid of Khufu, tried to systemize the process of providing food to workers.

In 14th century in China biometric authentication was rather popular among merchants . Technology of early biometrics was rather simple: paper with ink allowed to take palm print s and footprints of children in order to differentiate them from other.[3]

By the mid-1800s, with the rapid growth of cities due to the industrial revolution and more productive farming, there was a formally recognized need to identify people. Merchants and authorities were faced with increasingly larger and more mobile populations and could no longer rely solely on their own experiences and local knowledge. Influenced by the writings of Jeremy Betham and other Utilitarian thinkers, the courts of this period began to codify concepts of justice that endure with us to this day. Most notably, justice systems sought to treat first time offenders more leniently and repeat offenders more harshly. This created a need for a formal system that recorded offenses along with measured identity traits of the offender.

True biometric systems began to emerge in the latter half of the twentieth century, coinciding with the emergence of computer systems. The nascent field experienced an explosion of activity in the 1990s and began to surface in everyday applications in the early 2000s. [4]

?

3. Biometric Systems:

A biometric system is essentially a pattern recognition system that recognizes a person based on a feature vector derived from a specific physiological or behavioral characteristic that the person possesses as shown in Figure 1 [17]

3.1 Components of A simple biometric system:

1) Sensor module which acquires the biometric data; Web cam, Digitizing Table, Scanner.

2) Feature extraction module where the acquired data is processed to extract feature vectors; Projection [offline], DCT on Coordinates [online, offline]

3) Matching module where feature vectors are compared against those in the template ;( Neural Networks, Algorithm)

4) Decision-making module in which the user's identity is established or a claimed identity is accepted or rejected. (Final results) [18]

Biometrics As noted earlier refers to technologies that measure and analyze human body characteristics, such as DNA, fingerprints and eye retina and irises and hand measurements, for authentication purposes.

3.2 Biometrics Types:

1. Physical biometrics include this form of physical measurement and include methods such as face, fingerprints, iris scans, hand geometry, etc..

2. Behavioral biometrics are usually temporal in nature and include measuring the way in which the user performs certain tasks. This includes methods such as speech, signature, gait, keystroke dynamics etc..

3. Biometrics chemical [5] This is still nascent field and involves measuring chemical signals such as system and the chemical composition of the human race a number of biometric identifiers in use in different applications. Each biometric has its strengths and weaknesses and the choice depends usually on the application.

4. Biometric Systems Common Types:

This section will cover most common types of biometric system which are the following: fingerprint recognition, 3D face recognition, iris recognition, Voice recognition and hand recognition.

4.1 Fingerprint Recognition

This system, which consists of a hardware scanner and recognition software records specific fingerprint characteristics, saves each user data in a template. And then refers to the templates when the user next tries to gain access.

The fingerprint scanners assign a light through a prism that reflects off your finger to charge -coupled device (CCD). Creating an image that gets review Processed by an onboard computer it is important to note that the actual fingerprint image is not recorded. Instead, the devices perform a reduction of the image to data points called minutiae, which describe the fingerprint layout called a template as it explained in Figure2.

It works by matching relationships between minutiae: the points on your fingertips where print ridges end or divide

.

Human skin has two layers: epidermis and dermis. Dermis has also two layers: papillary and reticulated layer. In papillary layer find themselves in pairs pyramidal formations that are called papillary. Each pair of papillary is divided by channels of sweat glands. Such pairs make a row and covered by the layer of epidermis build comb of papillary lines. Papillary lines are situated chaotically but as streams. When three streams are near each other they build triangle which is called delta[3].

Papillary pattern is flexible. It means that there are no two similar papillary patterns in the world. Each person has its own unique papillary pattern. Each papillary pattern has its own unique details of structure: beginning and end of lines, merging and separation of lines, bends and breaks, ridges, eyes and hooks, breaks of papillary lines and oncoming places of their beginnings and ends.

The method of fingerprint is considered to be the most reliable method. The pluses of such method are: low cost of equipment, low time of procedure. But it has some minuses: papillary picture of the finger can easily be damaged; the system can be broken because of the high quantity of staff, some scanners �do not like �dry skin and it makes difficult for old people to use this method. The producers of scanners try their best to approve the quality of their goods.[6][7]

4.2 3D Face recognition

3D face recognition has attracted a lot of researches because it is a more reliable system and able to face facial expression and illumination problem.

The proposed method is illustrated by a block diagram representation in Figure 3 [8]

First depth map of all database facial images is threshold to discard their background using Otsu�s method. Then the resultant facial images are normalized to standard size 100x100 then aligned so that the nose tip is located in the center with the highest depth value 255 and all images depth values are between 0 and 255. Then, images are smoothed using Median filter with different window size to find best recognition rate and to prevent variation false matching results. After the smoothing stage, we train each smoothed individual image using 2DPCA where each image per individual is described by its principle component. The resultant principal components are used as feature vectors in the classification stage to calculate the similarity between two facial images as shown in Figure 4.[9] The nearest neighbor classifier is used in the matching process.

These methods cannot be compared because all of them use different scanners and databases. The ad-vantage of this method is that there is no need for contact, the low sensitivity to such factors as: beard, glasses, another form of hair, color of hair. Also 3D show high degree of reliability that can be compared with fingerprinting. But the negative side of the method is the expensive equipment, change of face expression reduces the statistical reliability of the method.

Facial recognition is not a perfect method of biometrics. As all other methods it has its own weak sides and strong sides. Dependence on the light, low resolution, sometimes form of hair, facial expression make the weak side of face recognition. The strongest side of the method is that it is not required aid from the test subject.

4.3 Iris recognition

Iris is a unique feature of the person. The basic characteristic visible of iris is trabecular network, that makes it possible to divide the iris in a radial manner. Formed in the eighth month of pregnancy. Iris is stable and does not change during the whole life[3].

The pattern of the iris (the band of tissue that surrounds the pupil of the eye) is

complex, with a variety of characteristics unique in each person. An iris recognition system uses a video camera to capture the sample and software to compare the resulting data against stored templates.

The process of iris recognition divided to three steps:

1. Image capture: image can be captured by a standard camera using infrared and visible radiation. Camera in the automated procedure are automatically locates the face and iris in the middle.

2. Locate the iris and image enhancement: when the iris is based, and iris recognition system determines eye only with better image focus and clarity. The image is analyzed. The purpose of this analysis is to determine the external border of the iris as the result of the analysis is the exact location of the circular iris

3. Store the image and compared: the division process, and filter segments of the iris to hundreds of vectors. The image is saved as � IrisCode and stored in the database[11].

The advantages of iris recognition is taken primarily by the iris of the eye to remain stable during the whole life. There is no connection between the user camera direct. The laser does not used, Just video technology. The high level of accuracy put the method in one row with fingerprinting. The method is remarkable for its high speed, scalability.

Although a large amount of benefits, and method also has some disadvantages. Iris is a small device and it is impossible to take the scanning process from a distance. For people with eye problems such as blindness, cataracts, it will be very difficult to participate in the recognition process because it is very difficult to read the iris. Without the right amount of lighting it is difficult to capture the image.

4.4 Voice recognition

Speech or voice-based recognition systems identify a person based on their spoken words. The generation of human voice involves a combination of behavioral and physiological features. The physiological component of voice generation depends on the shape and size of vocal tracts, lips, nasal cavities, and mouth.[13] Speaker recognition is highly suitable for applications like tele-banking but it is quite sensitive to background noise and playback spoofing. Again, voice biometric is primarily used in verification mode.

The proposed method will help for voice recognition where we take voice as input through microphone and then register for Online examination. Then at the time of logging user will logged in to the online examination system through voice authentication, which uses FFT for comparing input voice of the user with the template voice. Next the user will proceed with the questions displayed and read by the system..[12]

A- Voice Recognition

Speech recognition is the process by which a computer (or other type of machine) identifies spoken words. Basically, it means talking to your computer, and having it correctly recognized what you are saying. For the voice recognition part the following steps have to be followed:

i) At first, Provide the user details as input voice asked by system.

ii) The system will then generate a �.wav� file and stored it in the DB.

iii) At the time of log in by the user, user needs to provide the same information given at the time of registration and the system compares the recorded voice with the one saved in database. If both match, user logs in successfully, otherwise not.

B- Comparison of voice recordings.

Comparison the Voices entered by user X1,X2 by the storing two voices in .wav files Y1,Y2. Then we plot both signal and try to match them, Using mathematical functions we compute and plot an average power spectrum [9] which is also normalized to compare it with two individual voices giving us desired results as shown in Figure 6,7

.

C- Authentication of voice:

For authentication of voice an efficient spectrum analysis to continuous-time signal based on DFT, the Fast Fourier Transform (FFT) method is adopted. It breaks down DFT of N-points sequence to the one of shorter sequence gradually, which will greatly improve the calculating speed of DFT.

A fast Fourier transform (FFT) is an efficient algorithm to compute the discrete Fourier transform (DFT) and its inverse.

In the context of Fast Fourier transform algorithms, a butterfly is a portion of the computation that combines the results of smaller discrete Fourier transforms (DFTs) into a larger DFT, or vice versa (breaking a larger DFT up into sub transforms)[12].

For recognition the system match the frequency spectra of recording and frequency spectra of average as in the figure 8.

The most significant difference between voice biometrics and other biometrics is that voice biometrics are the only commercial biometrics that process acoustic information. Most other biometrics are image-based. Another important difference is that most commercial voice biometrics systems are designed for use with virtually any standard telephone on public telephone networks. The ability to work with standard telephone equipment makes it possible to support broad-based deployments of voice biometrics applications in a variety of settings. In contrast, most other biometrics require proprietary hardware, such as the vendor�s fingerprint sensor or iris-scanning equipment.

?

4.5 Hand recognition

This method is simple, easy and inexpensive. It has been established in many locations in the world. Hand geometry recognition systems are based on a number of measurements taken from the human hand. It measures the shape of hand, the size of palm, and the lengths and widths of the fingers. Moreover the size of the hand is big and it is not currently in wide deployment for computer security applications primarily because it requires a large scanner [15].

The main characteristics for this method are measuring and recording the height, length of the fingers, distance between joints, shape of the knuckles, surface area of the hand[16]

This system explores the use of hand geometry as a measure of a person's identity. The system consists of an acquisition device that captures the top view and side view of a user's right hand as he places it on the flat surface of the device, Hand and finger geometry are shown in Figure. 9

As summary of above 5 types of biometric systems, the tables1,2[17] will explain more about them.

5. Attacks against biometric systems:

Although biometric systems have many advantages over symbolic traditional (for example, key) or knowledge (for example, a password) systems based authentication (for example, increase user comfort and durability against users quorum),but they are still vulnerable to attacks. Figure 10 [20] shows the locations of these attacks in a generic biometric system.

5.1 The Motivations of attacking on a biometric system:

1.When a person wants to hide his own identity

2. Because everyone wants to make concessions to the other person.

3. The benefits coming from sharing biometric trait.

5.2 The locations of these attacks in a generic biometric system: as shown in Figure 10.

1. Could be a trick directly, such as fake finger, and a copy of the signature, or a face mask.

2. This kind of attack, is its response to the data previously recorded in the biometric system to digital system is penetrated himself, and therefore also called as "replay attack", for example, provide a digital copy of the fingerprint image or recording audio signal from the speaker.

3. The feature extractor may be attacked with a Trojan horse program that produces predetermined feature sets.

4. Legitimate feature sets extracted from the biometric input may be replaced with synthetic feature sets. For example, if minutiae of a fingerprint are transmitted to a remote matcher (say over the Internet) than this threat is very real.

5. The matcher may be attacked with a Trojan horse program that always directly produce a specified result - match, no match, or a score.

6. Could be an internal attack are modified templates registered in the database, remove, or can the introduction of new models in the database, which could lead to permission for a fraudulent individual, or at least denial-of-service for a person who is associated with the template is damaged

7. The enrolled templates in the stored database are sent to the matcher through a communication channel which could be attacked to change the contents of the templates before they reach the matcher.

8. The final decision by the biometric system is bypassed or replaced by another decision with the choice of produce for the pirates. Even if feature extraction and matching units performance characteristics have had excellent, has issued the feasibility of a simple exercise to overcome it.

5.3 Spoof attacks (it is called direct attack): An impostor may attempt to spoof the biometric trait of a legitimate enrolled user in order to circumvent the system [21]. This type of attack is mostly expected when behavioral traits are used such as signature and voice. However, physical traits are also susceptible to spoof attacks. For example, it has been demonstrated that it is possible to present an image (i.e. a photo or upon the notebook�s display) to circumvent a face verification system[21]

The Spoof attacks like Fingerprint spoofing as shown in figure 11 [19]

5.4 Known Technologies To Resist the Attacks

5.4.1 Liveness Detection Mechanisms

Detect Liveness to thwart attacks at the point of attack: detection can be implemented using software or hardware like the following:

� the use of additional hardware for signs of life, such as temperature, pulse detection, blood pressure and other fingerprint and facial movements. A disadvantage is that additional hardware makes the system expensive and bulky using information that was already captured to detect signs of life.

� Use information Liveness inherent in biometric obtained. For fingerprints, and can use the impression the near side of the nail, The system can be used using multiple instances of the same biometric revealed by asking Liveness for the user to provide a random subset of biometric measurements, for example on the left index finger and then right middle finger.

� The research by the laboratory biomedical at the University of West Virginia on an algorithm based on detection of perspiration in a time progression of fingerprint images.

5.4.2 Steganographic and Watermarking Techniques

"Steganographic and Watermarking techniques are used to resist attacks at the attack points 2 and 7 (Channel between the sensor and feature extractor and also the channel between the stored template and the matcher)"[23]. Hide confidential information communication sense, it involves hide critical information in the data carrier unsuspected. Information can be hidden based on techniques that are appropriate for the transfer of vital information important from a client to a server. . This method is used to secure biometric authentication systems for commercial transactions against replay attacks. To achieve this, the service provider issued a series verify different for each transaction. Series are mixed with fingerprint image before sending. When you receive the picture by the service provider is pressure, and the image is examined for verification chain for once. Here, the message is hidden in a fixed location, but is deposited in different places on the structure of the picture so that cannot be recovered easily. . Watermark on the occasion of the information in the database of biometric template allows the integrity of the contents to be verified when retrieving the matching..

5.4.3 Multi-modal Biometric Systems

Multi-modal biometric systems can be used to resist spoofing attacks (attacks at point 1). Difficult for a hacker to biological traits parody at the same time from multiple legitimate user. Choice and decided a number of biological traits according to the nature of the application, the computational demands and input costs, and the relationship between the qualities considered.

The noise problem can be reduced in the data obtained by using multi-modal biometrics and identify different degrees of importance for different attributes. This, in turn, leads to better performance and accuracy of matching that makes spoofing attacks more difficult. Since the multi-model biometric system increase the computational overhead and costs, it should be examined cost versus performance trade-off prior to the deployment of these systems.

5.4.4 Soft Biometrics

Soft biometrics can be used to thwart attacks at the attack points 1 and 8 (attacks on the sensor and decision maker). Soft biometric traits are those characteristics that provide some information about the individual, but lack the differentiation or always to distinguish enough any two individuals (sex, race, age, height, weight, etc.)[23]. Most biometric data collection systems to help users during the recording, which is stored either in the database or smart card owned by the user. The information that has been collected to help along with dozens of matching lead to the correct identification of the user, and this in turn prevents theft. Factors such as age, sex, color, etc. affect the performance of identity verification system. Use soft biological traits help to filter large biometric database for a limited number of templates to do compared to, and that, in turn, will improve the speed and efficiency of the identity verification system.

The table 3 [23] bellow list the advantages and drawbacks of the above techniques

Drawbacks Advantages Technique

Increased cost for the extra hardware and software, user inconvenience and increased acquisition time. Resists spoofing attacks. Liveness Detection

Problem of image degradation and lack of algorithms to deal of algorithms to deal with it. Prevents replay attacks and provide integrity of the stored templates. Watermarking

Lack of techniques for automatic extraction of soft biometric techniques. Provides improved performance through filtering and tuning of parameters. Soft Biometrics

Increased system complexity, computational demands and costs. Improves performance, resists spoofing and replay attacks and provides high population coverage. Multi-modal

?

6. Finger print as a Case study:

Choosing the finger print that it is popular which the people interested to use it instead of face or hand recognition that may be effect the health of them as they think, also it is stable and can depend on it as well as the safe way and more accurate is to use the finger print to reduce the problem. A major advantage of finger imaging is the long time use of fingerprints and its wide acceptance by the public and law enforcement communities as a reliable means of human recognition

6.1 Related Works (Previous Studies):

The starting of � Finger Prints� in 1892 with Sir Francis Galton publish his book �where three main fingerprints patterns were described : loops, whorls , arches. It should be pointed out that he offered to use fingerprints from all 10 fingers.

The upgrades on this method by the time until 1992 when the immigration system used fingerprints for the first time[4].

Brasilia began to issue biometric passport at the end of 2010, only in the capital area and Goiaz state. But it took one year and biometric passports were all over the country. Passports contain the common set of personal information, digital photo, finger-prints.

Canada before the end of 2012 will have ePassports which will incorporate biometric passport. There will be a chip, with all personal information, as well as information found in machine -readable zone, China began to use biometric passports in 2011. Passport contains digital photo of the owner, fingerprints and all other biometric features of the holder[3].

From table 1,2 in the section 4 , The safe way, the cheapest, the easiest, the standard one and more accurate is to use the finger print�

6.2 Finger Prints System Overview:

Fingerprints are usually considered to be unique, with no two fingers having the exact same dermal ridge characteristics.

6.2.1 Classifying Fingerprints

1- Arch papillary pattern. They are simplest in their structure and according to the frequency of meeting � 5%. They consists of not more than two streams of papillary lines that starts at one side end and go to another end , making in the middle of the pattern arcing figures.

2- Papillary pattern. This type is the most popular; about 60%-65% of people have this type of pattern. The picture is built by three streams of lines

3- The third type is whorl, is met at about 30% of people. The inner picture can be made by papillary lines as ovals, spirals, loops, or their combinations.[3]

6.3 System Operations:

Optical, silicon and ultra sound are the main technologies used to capture the fingerprint images with sufficient detail.

6.3.1 Process of fingerprint analysis

1. Scanning of a fingerprint image.

2. Image quality improvement.

3. Image processing.

4. Algorithms to recognize fingerprints: Minutia matching compares specific details within the fingerprint ridges. Pattern matching compares the overall characteristics of the fingerprints.

5. Authentication

a. User Registration In any secure system, to enroll as a legitimate user in a service

b. Fingerprint Enhancement A fingerprint is made of a series of ridges and furrows on the surface of the finger

6.4. System Requirements:

This part will cover the kinds of hardware will be used and the reason to use that hardware, also the kinds of software will be used and the reason touse that software.

6.4.1 Hardware Architecture:

There are 6 types of sensors for fingerprint recognition:

1- Optical Fingerprint Sensors

2- Thermoelectric Sensors

3- Capacitive Sensors

4- E-Field Sensors

5- Touchless Sensors

6- Surface Pressure Sensor

The Hardware requirement also is the network equipment like switches and Adaptors, and Pc devices as shown in the figure 14

6.4.2 Software Architecture

choosing C# and Microsoft Access DB for my design software.

Dot Net is easy to use and popular in the design area. Microsoft Access DB also can help us to make the DB easy to design and update, The number of records is limited, so the best choose is Access DB.

But for the end user interface I choose macromedia flash it is easy to use and popular in the design area. The second point is I don�t want to have a static and boring interface. So, I decided to use flash animation to replace of static picture interface.

7. Finger Print Biometric System user interface:

I have two Interfaces at all one with the visitor who use the machine to be authorized user and the system Interface which used by Administrators and HR employees.

7.1 Visitor Interface:

7.2 Admin Interface:

1- The Main window

?

2- The devices window

3- The users window

?

4- The link between users and the devices.

5- The Monitoring window

?

8. Design the Data Flow and the database structure of biometric systems.

8.1 Design DFD Diagram:

?

8.2 ER Diagram

8.3 DB Design

8.4 Database structure

1- UsersMachines

a. ID AutoNumber , Primary Key

b. USERID Number, Foreign Key (USERINFO -> USERID )

c. DEVICEID Number, Foreign Key (Machines -> ID )

2- Machines

a. ID AutoNumber , Primary Key

b. MachineAlias Text

c. ConnectType Number

d. IP Text

e. Port Number

f. Enabled Yes/No

g. CommPassword Text

h. usercount Number

i. fingercount Number

j. LockControl Number

k. Purpose Number

3- CHECKINOUT

a. USERID Number

b. DeviceDef Number, foreign Key (usersMachines -> ID )

c. CHECKTIME Date/Time

d. CHECKTYPE Text

e. VERIFYCODE Number

f. SENSORID Text

g. LOGID AutoNumber, Primary Key

?

4- Userinfo

a. USERID AutoNumber, Primary Key

b. NAME Text

c. GENDER Text

d. TITLE Text

e. BIRTHDAY Date/Time

f. HIREDDAY Date/Time

g. STREET Text

h. CITY STATE Text

i. ZIP Text

j. OPHONE Text

k. FPHONE Text

l. DEFAULTDEPTID Number

m. PASSWORD Text

n. PHOTO OLE Object

o. CardNo Text

5- DEPARTMENTS (used in drop down list)

a. DEPTID AutoNumber

b. DEPTNAME Text

9. CONCLUSION:

� Biometric systems will in the future become more famous in civil working of many areas. Probably will in a few years to the arrival of those granted a private home or a car on a successful scan the iris, which makes the house or car keys traditional.

� Discussed in this literature range of potential attacks and one of the most important of these attacks is known as "a spoof attack" (also called "direct attack" since it is carried out directly on the biometric sensor). Spoof attacks are of great importance to a process that does not require advanced technical skills, and therefore, the potential number of attackers is very large. Diversity must therefore verification tools more than kind to avoid such risks

� Fingerprint system properties qualify to be a universal model in both institutions and companies, especially those that has no big budget to buy major systems to manage this aspect of confidentiality and verification.